Presentation Details
| Closing the loop between AI and animal behavior Sandra Winters. University of Colorado Boulder, Boulder, CO, USA |
Abstract
Machine learning and artificial intelligence (“AI”) are powerful approaches for studying the evolution of animal behavior and morphology, especially when carefully grounded in the underlying biology of modeled systems. I discuss the advantages and drawbacks of AI-based research, highlighting research in animal communication as a representative case study. I emphasize the importance of careful calibration and validation of AI-generated results in biological systems as key to avoiding potential pitfalls, and point to AI and experimentation as complementary approaches that can reinforce one another: AI can search for patterns or (potential) phenotypic optima that would otherwise be nearly impossible to identify, and experiments can demonstrate causality and ground AI-generated results in living systems. To unlock the full potential of AI, future animal behavior research should embrace an integrative and multidisciplinary approach that harnesses the power and scope of AI alongside traditional methods that are critical for evaluating predictions in naturally behaving animals.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
Content Locked. Log into a registered attendee account to access this presentation.